DNS Issues after making VPN Public
We are running into a problem where SOMETIMES our users connect to the VPN and all webpages fail to load. I can confirm this is a DNS problem. As soon as we change the DNS on the network adapter, it works just fine. This is also intermittent. Sometimes I can have users just disconnect and reconnect until it works.
Our devops team recently disabled split tunnelling for this VPN in order to create a public static IP address for it. I have a hunch this is related.
I found a promising article here but was not sure how to "prioritize". https://www.reddit.com/r/aws/comments/q39ji2/aws_client_vpn_no_local_internet_when_splittunnel/
Our routing table is as follows:
When we connect to the VPN, I can confirm that the DNS DOES change to the DNS server in AWS.
Any help is appreciated!
- Newest
- Most votes
- Most comments
It looks like all your routes are bounded to one specific subnet only.
Do you happen to have more than one target network associated with your Client VPN endpoint?
If this is the case, then you are missing routes on your other subnet(s) which can explain the intermittent problem when your client is connected to the subnet with the missing routes.
Have a look at this link.
Make sure you have the DNS Server settings configured on the client VPN Endpoint https://repost.aws/knowledge-center/client-vpn-how-dns-works-with-endpoint
Relevant content
asked 2 months ago
asked 4 years ago
- AWS OFFICIALUpdated a year ago
- AWS OFFICIALUpdated 3 years ago
